Perinatal conditions account for 60% of US neonatal deaths, yet little is known about rates of morbidity attributable to these conditions. To estimate these rates, we analysed newborn hospital discharges from the National Hospital Discharge Survey. We used International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes to classify discharge diagnoses among a weighted, nationally representative sample of newborns discharged from short-stay, non-federal US hospitals. We compared overall and cause-specific morbidity rates attributable to perinatal conditions (ICD-9-CM 760.0-779.9), as well as the average length of hospital stay among newborn discharges during 1989-90 and 1999-2000. The overall newborn morbidity rate declined from 36.3% in 1989-90 to 33.7% in 1999-2000 (P < 0.01), despite significant increases in high-risk births. The decline can be attributed to significant decreases in the reported rates of jaundice, fetal distress, birth trauma and birth asphyxia. Rates of jaundice decreased from 15.7% to 13.4% (P < 0.01). The average length of stay decreased among newborns with no morbid condition (2.37-2.04 days, P < 0.001) and among those with one perinatal condition (3.11-2.51, P < 0.001), but increased among those with multiple perinatal conditions (8.43-9.98, P < 0.05). Morbidity rates among newborns discharged from US hospitals declined. Shorter newborn hospital stays may have resulted in fewer cases of jaundice being diagnosed before discharge. Stricter diagnostic criteria and changes in obstetric practices may have led to a decline in the rates of fetal distress, birth trauma and birth asphyxia.